Background-Atrial fibrillation (AF) is a major cause of stroke. Although standard investigations after an event include electrocardiographic monitoring, the optimal duration to detect AF is unclear. We performed a systematic review and meta-analysis to determine whether the duration of electrocardiographic monitoring after an ischemic event is related to the detection of AF. Methods and Results-Prospective studies that reported the proportion of new AF diagnosed using electrocardiographic monitoring for >12 hours in patients with recent stroke or transient ischemic attack were analyzed. Studies were excluded if the stroke was hemorrhagic or AF was previously diagnosed. A total of 31 articles met inclusion criteria. Longer duration of monitoring was associated with an increased detection of AF when examining monitoring time as a continuous variable (P<0.001 for metaregression analysis). When dichotomizing studies based on monitoring duration, studies with monitoring lasting <= 72 hours detected AF in 5.1%, whereas monitoring lasting >= 7 days detected AF in 15%. The proportion of new diagnosis increased to 29.15% with extended monitoring for 3 months. Significant heterogeneity within studies was detected for both groups (<= 72 hours, I-2=91.3%; >= 7 days, I-2=75.8). When assessing the odds of AF detection in the 3 randomized controlled trial, there was a 7.26 increased odds of AF with long-term monitoring (95% confidence intervals [3.99-12.83]; P value <0.001). Conclusions-Longer duration of electrocardiographic monitoring after cryptogenic stroke is associated with a greater detection of AF. Future investigation is needed to determine the optimal duration of long-term monitoring.